Sage Journals: Discover world-class research

Abstract

Immunoassays have long been widely used in a variety of applications, such as for medical diagnostics, pharmaceutical analysis, environmental, food safety testing, and for basic scientific investigations because of its simplicity, sensitivity, and specificity. Microfluidic systems, also well known as a “lab-on-a-chip” or a “micro-total-analysis-system” have attracted a lot of attention in the past two decades because of advantages associated with miniaturization, integration, and automation. A promising platform for the combination of these two technologies, microfluidic immunoassays, has been extensively explored in recent years. The aim of this article is to review recent advancements in microfluidic immunoassays. A brief introduction to immunoassays and microfluidic devices will include a literature review, followed by an in-depth discussion of essential techniques in designing a microfluidic-based immunoassay from different perspectives, including device substrates, sample/reagent transportation, surface modification, immobilization, and detection schemes. Finally, future perspectives on microfluidic immunoassays will be provided. These developments with microfluidic immunoassays may provide a promising tool for automatic, sensitive, and selective measurements in practical applications.

Keywords

microfluidics immunoassay immobilization lab-on-a-chip MEMS

Introduction

Immunoassays

Antibodies (Abs) are proteins produced in animals and human bodies by immunological responses to the presence of allochthonous substances called antigens (Ags). They have a highly specific affinity for these Ags in nature. Each Ab has a unique structure recognized by a corresponding Ag in a lock-and-key mechanism. Immunoassays have a variety of formats, all of which make use of the sensitivity and specificity of this Ab—Ag interaction, which allows for the quantification and monitoring of small molecules, such as drugs and metabolites,¹ large proteins,² nucleic acids,³ and even whole pathogens.⁴ They have been widely used in clinical analysis, food safety and environmental monitoring, and basic bio-technological investigations.

Immunoassays can be used to detect either Abs or Ags according to the needs of the experiments. Generally, immunoassays can be classified as either a competitive or a noncompetitive format.⁵ In the competitive format as shown in Eqs. (1) and (2), the unlabeled Ags competes with labeled Ags (Ag*) for a limited number of Ab-binding sites. As the amount of unlabeled Ags in a sample increases, the amount of labeled Ags bound to the Abs decreases, resulting in a decrease in the detection signal if the Ab-bound Ags (Ag*—Ab) are detected, or an increase in signal, if the labeled free Ags are detected.

Ag + ab \leftrightarrow Ag - ab

(1)

Ag * + Ab \leftrightarrow Ag * - Ab

(2)

Alternatively, in the noncompetitive format, as shown in Eq. (3), Ags in a sample conjugate with an excessive amount of labeled Abs (Ab*) to form a complex that is strongly dependent on the number of Ags, resulting an increase in the detection signal as the Ags in the sample increases.

Ag + Ab * (excess) \leftrightarrow Ag - Ab * + Ab * (excess)

(3)

Immunoassays can also be divided into heterogeneous and homogeneous formats.⁵ In a heterogeneous format, Abs or Ags are immobilized on a solid substrate where the complex forms. On the other hand, in a homogeneous format, the conjugation takes place in the solution phase. Heterogeneous immunoassays take advantage of the high surface area/volume ratio and result in good performance in sensitivity. However, inextricable steps are sometimes required for the immobilization of Abs or Ags on the solid substrate. Homogeneous immunoassays benefit from the multiplexing and fast electrophoretic separations. However, preconcentration steps are usually required because the surface area of the substrate is not used for immobilization of Abs or Ags. Both formats have been extensively studied and can be easily implemented in microfluidic devices. This section only provides a brief introduction to immunoassays. Detailed information about immunoassays can be found in the literature.^{6 –8}

Microfluidics

Microfluidic systems fabricated by microelectromechanical systems (MEMS) technology are now usually referred to as“lab-on-a-chip” (LOC),“biochips,” or“micro-total-analysis-system.” They are often envisioned as miniaturized versions of their large-scale counterparts. These miniaturized systems can carry out entire protocols traditionally performed in a laboratory. Sample pretreatment, sample/reagent transport, mixing, reaction, separation, detection, and product collection can all be performed automatically on a single LOC system. Functional microfluidic devices, such as micropumps, microvalves, microfilters, microreactors, and microseparators can be microfabricated and even integrated to perform a specific assay. The advantages of these developed LOC systems include less sample/reagent consumption, a reduced risk of contamination, enhanced sensitivity, less unit cost, lower power consumption, and a higher reliability and functionality. More importantly, portability arising from their compact form is a key factor for point-of-care (POC) applications. Despite these advantages, there are still potential limitations, such as bubbles formation and dead volume in microfluidics that need to be addressed. The bubbles formed in microdevices disrupt the continuity of the liquids and may result in poor performance. This is particularly serious in electrophoresis-based microdevices because the bubbles disturb the applied electric field.⁹ Dead volume is another key issue in microfluidics. In the microscale environments, dead volume may cause serious contaminations and hence poor results.¹⁰ The works in LOC have grown rapidly in the past two decades, and many review articles are available.^{11 –15} In this article, only recent advancements in microfluidic immunoassays are covered here.

Immunoassays in Microfluidic Systems

The process in most immunoassays includes a series of washing, mixing, and incubation steps, which are labor intensive and time consuming, which often takes several hours, sometimes even up to 2 days to perform one single assay. Most of the time required in a long immunoassay is mostly because of the long incubation time attributed to inefficient mass transport for the immunoagents to move from a solution to the surface where the conjugation occurs because the immunoreaction itself is relatively rapid.¹⁶ Moreover, the immunoagents used in immunoassays are relatively expensive. The consumption of the immunoagents can be greatly reduced if the system is miniaturized. Therefore, there is a demand to develop an automated and miniaturized platform for immunoassays. Such a platform must be capable of simplifying procedures, reducing the assay time and sample/reagent consumption and enhancing the reaction efficiency. The advantages of the microfluidic systems described previously fulfill these important criteria for immunoassays. Therefore, extensive investigations using microfluidics for performing immunoassays have been reported recently. The following sections introduce some representative original articles in microfluidic immunoassays published between the years 2005 and 2009. It should provide readers a comprehensive understanding of this promising technology.

Substrate Materials for Microfluidic Immunoassays

Microfluidic devices for immunoassays can be fabricated from a variety of materials. The most commonly used substrate materials are silicon, glass, and polymers. Each of these materials has its own advantages and limitations. In this section, a brief introduction to these materials is first given and a comparison of material properties is listed in Table 1.

Table 1.

Comparison of silicon, glass, and polymer as materials for microfluidic immunoassays

	Silicon	Glass	Polymer
Development period	Early	Later	Latest
Surface chemistry	Well developed	Well developed	Well developed
Thermal conductivity	Good	Good	Poor
Endurance to high temperature	Good	Good	Poor
Structure aspect ratio	Low (may be high if dry etching has been adopted)	Low (may be high if dry etching has been adopted)	High to low
Optically transparent a	No	Yes	Yes
Electrical conductivity	Yes	No	No
Substrate cost	Expensive	Less expensive	Low cost
Fabrication cost	Expensive	Less expensive	Low cost
Mass productionb	Yes	Yes	Yes

Although glass and polymers are optically transparent, they may adsorb light in the UV spectrum.

Although all the materials are capable of being mass produced, the fabrication processes for silicon and glass are relatively slow. For polymers, replication methods for PDMS are not as convenient as those used for PC, PS, and PMMA in practical and commercial aspects.

Silicon

Silicon is used as a material for fabricating microfluidic devices because of the well-established microfabrication processes developed by the microelectronic and the MEMS industry and from extensively studied surface chemistry. Furthermore, it possesses good thermal conductivity and is resistant to high temperatures; therefore, is suitable for applications requiring a relatively high operating temperature, such as for a polymerase chain reaction (PCR) and for bioreactions. The most popular fabrication method for silicon-based microfluidic structures is wet etching,¹⁷ usually resulting in microstructures with low aspect ratio features that may be a limitation in most applications. This issue may be addressed by using dry etching,¹⁸ which is a relatively expensive microfabrication process. Moreover, silicon is not optically transparent and is electrically conductive, thus resulting in limitations for optical and electrochemical detection and for electrokinetic transportation, which are key issues with microfluidic immunoassays. More importantly, silicon substrates are relatively expensive when compared with other materials, such as glass and polymers. Furthermore, the fabrication process for silicon-based microfluidic devices involving substrate cleaning, resist coating, photolithography, development, and wet/dry etching is relatively time consuming and costly. These limitations hinder its practical applications in commercial immunoassays.

Glass

In contrast to silicon, glass substrates are less expensive, optically transparent throughout the visible spectrum, and not electrically conductive. These properties overcome some of the problems associated with silicon-based microfluidic devices. Therefore, glass seems to be an ideal substrate for performing microfluidic immunoassays. However, glass-based microfluidic devices are usually fabricated by a wet etching method,¹⁹ which still limits the etched microstructures to low aspect ratio features. Again, some expensive dry-etching processes for glass can be applied to address this limitation.²⁰

Polymers

Polymers are alternative materials to silicon and glass in microfluidic immunoassays because of their relatively low cost and simple fabrication process. Many polymers have been used in microfluidic applications, such as polymethyl methacrylate (PMMA),²¹ polycarbonate,²² polystyrene (PS),²³ and polydimethylsiloxane (PDMS).²⁴ Among these polymers, PDMS is one of the most commonly used materials for microfluidic immunoassays in recent studies because of its desirable characteristics, such as flexibility, optical transparency (down to 230 nm), and biocompatibility. PDMS is a polymer composed of repeating OSi(CH₃)₂ monomers, resulting in a hydrophobic surface, which is the major limitation when used with immunoassays because of the nonspecific adsorption of proteins and other molecules, poor wettability with aqueous solutions, and the formation of bubbles.²⁵ This disadvantage, however, can be easily overcome by bulk/surface modification and well-developed functionalization techniques.^{26 –31}

Heterogeneous Substrate Materials

Recently, heterogeneous materials formed by using two different types of materials for microfluidic devices have been extensively explored. For example, PDMS/glass³² and silicon/glass³³ are commonly used for forming microfluidic devices. These hybrid microfluidic devices provide advantages attributed to their component materials. For example, a microchip made of PDMS and glass can fulfill the requirements of fast fabrication, low cost, flexibility, biocompatibility, and good thermal conductivity.

Sample/Reagent Transport

Liquid transport plays an important role in microfluidic immunoassays because the performance of the liquid driving system can directly affect the results of the assays. Many strategies for on-chip liquid transport in microfluidic immunoassays have been reported in the literature, such as electroosmotic,^34,35 electrowetting,^{36 –38} pneumatic,^{39 –42} centrifugal,^{43 –48} power-free,⁴⁹ piezoelectric,⁵⁰ and thermopneu-matic⁵¹ approaches. A comparison of these on-chip liquid transport strategies is listed in Table 2. External power sources, valve integration, and flow rates for each pumping method are also listed.

Table 2.

Comparison of strategies for on-chip sample/reagent transportation in microfluidic immunoassays

	External power source	Valve integration	Flow rate	Refs.
Electroosmosis	Electrical voltage	Yes	nL/s toμL/s	34, 35
Electrowetting	Electrical voltage	Noa	NAb	36–38
Micropump	Compressed air pump	Yes	μL/min to mL/min	39–42
Centrifugal force	Rotary motor	Yes	nL/s to submicroliter/s	43–48
Power-free	Not necessary	Yes	nL/s	49

Although on-chip valves are not integrated in the electrowetting-based transport method, liquids can be immediately stopped when the electric voltage is not applied.

The flow rate for electrowetting is normally expressed as a flow velocity, and ranges from 0.01 to 0.1 m/s for a droplet with a radius less than 10⁻³ m.

Electroosmotic Pumping

Electroosmotic pumping based on electroosmotic flow (EOF)³⁴ is a commonly used liquid transport strategy in microfluidic systems because of the ease of automation and no moving parts are required. EOF is the bulk flow resulting from the effect of the electric field on the solution double layer at the channel wall. Counterions (typically, cations) building up near the wall surface to maintain charge balance form the double layer and create a potential difference very close to the wall surface, which is known as the zeta potential. The zeta potential is determined by the surface charge on the channel wall. Because this charge is strongly pH dependent, the magnitude of the EOF varies with pH. The zeta potential is also dependent on the ionic strength of the buffer. Increased ionic strength results in double-layer compression, decreased zeta potential, and hence reduced EOF. As the electric field is applied, the cations forming the diffuse double layer are attracted toward the cathode. Using this mechanism, the steps for flow switching, sequencing, and stopping can be easily achieved by controlling the applied electric field. For example, Gao et al.³⁵ developed a microchip for simultaneous detection of multiple microbial Ags based on electroosmotic pumping. As shown in Figure 1, Ags were first adsorbed onto a PDMS-coated glass slide by using a microfluidic network, followed by bonding to another PDMS slab-bearing H-shaped microchannels designed for delivering solutions of the primary Ab, the washing buffer, and the secondary Ab. The sequential pumping of these solutions was achieved by using a programmable high voltage sequencer equipped with eight independent outputs. Using this method, the determination for Escherichia coli (E. coli) lysate Ag has been achieved with a detection limit of 3μg/mL. Moreover, the simultaneous detection of E. coli and Helicobacter pylori (H. pylori) lysate Ags has also been achieved. This microfluidic system represents the potential for the simultaneous detection of multiple pathogenic infections.

Figure 1.

The fabrication process of the microchip for simultaneous detection of multiple microbial antigens based on electrokinetic pumping. (A-B) Antigen immobilization and (C) bonding with a PDMS slab-bearing H-shaped microchannels and forming the reaction regions. Reprinted from Ref. 35 with permission from Springer Science+Business Media (Copyright 2009 Springer Science+Business Media).

Electrowetting Force

Electrowetting is a liquid transport strategy whereby liquid droplets are manipulated in the presence of programmed voltage sequences applied to an electrode array.³⁶ It can be a promising platform for performing immunoassays. For instance, combining electrowetting and magnetic microbead-based immunoassays, Sista et al.³⁷ demonstrated a microfluidic chip for performing heterogeneous immunoassays. As shown in Figure 2, a sample droplet and a reagent droplet containing capture Ab-conjugated magnetic beads, reporter-conjugated Abs, and blocking proteins were dispensed on the surface of the chip. These two droplets were then merged, mixed, and incubated by electrowetting-based transport, followed by applying a magnetic field to capture the sandwiched microbead complexes. Then, unbound components were split from the merged droplet, followed by the introduction of the washing buffer. The splitting and washing steps were repeated several times to thoroughly remove unwanted components. Finally, a reagent droplet was added for chemiluminescent detection. Using this electrowetting system, a sandwiched heterogeneous immunoassay on human insulin and interleukin-6 (IL-6) has been successfully performed within 7 min. This electrowetting-based microfluidic platform can be also successfully used for the identification of a cardiac marker, troponin I, from whole blood samples,³⁶ sample preparation for bacterial pathogens indicating infectious diseases,³⁶ and the extraction of human genomic DNA.³⁸ It is believed that the electrowetting-based microfluidic immunoassay is adaptable for POC applications because of its relatively low cost and simple instrumentation.

Figure 2.

Merging (A), incubation (B), immobilization (C), splitting (D), and washing (E) processes in an electrowetting-based immunoassay. Reprinted from Ref. 37 with permission from the Royal Society of Chemistry (RSC) (Copyright 2009 The Royal Society of Chemistry).

Micropneumatic Pumping

The micropneumatic pump is composed of a liquid flow microchannel, flexible PDMS membranes, and air chambers to pneumatically deflect the PDMS membranes in sequence via the introduction of compressed air.^{39 –42} As shown in Figure 3, the peristaltic effect driven by the time-phased deflection of PDMS membranes along the microchannel can be used to deliver liquid in the microchannel. It has been demonstrated that micropneumatic pumps, incorporated with microvalves, can be used to determine the presence of Abs specific to hepatitis C virus and syphilis from serum samples.³⁹ In this work, the Ag was first coated on the surface of a detection area. Then, serum samples, a washing buffer, a horseradish peroxidase (HRP)-labeled secondary Ab, a developing buffer, and a stopping buffer in individual reservoirs were sequentially transported to the detection area by the micropneumatic pump, followed by optical detection via the measurement of absorbance. Usually it takes at least three electromagnetic valves to control the sequential deflection of the PDMS membranes. Alternatively, similar micropneumatic pumps have been reported, in which different-sized air chambers were interconnected by air channels to transport liquids in microchannels.⁴⁰ Because the interconnecting air channels allowed for the sequential actuation of the air chambers, this generated a peristaltic-like activation of the PDMS membranes, only one electromagnetic valve was required to perform the pumping. By using this micropneumatic pump, bead-based flow cytometry using immunoassays was performed in a microchip.⁴¹ The delivery of virus samples, a washing buffer, virus-specific-Ab-conjugated magnetic beads, dye-labeled Abs, and sheath flows for focusing the sample stream were all transported by the micropneumatic pump. Using this micro flow cytometer chip, the entire process, including sample/reagent delivery, mixing, incubation, washing, detection, and the collection of beads was completed within 40 min. Recently, another micropneumatic pump was reported, which was incorporated with a normally closed valve for generating high pumping rates and to prevent any backflow from the destination wells after releasing the compressed air.⁴² Using this liquid transport method, it avoided contamination from the waste chambers, and this design is expected to be applicable for many future microfluidic applications.

Figure 3.

The sequential deflection of PDMS membranes in a micropneumatic pump.

Centrifugal Force

The centrifugal force has been used for liquid transport in compact disc (CD)-based immunoassays in recent years, and some review articles are available.^43,44 Immunoassays on a CD are a promising platform to realize many POC applications by embedding many analytical functions performed in a laboratory, such as sample/reagent transport, metering, dilution, mixing, separation, and detection. All these functions can be performed automatically in a disc format. It has been expected to be a powerful platform for medical and clinical diagnostics. It has great commercial potential because of simple instrumentation, ease for economic mass production, and easy adaptation to existing detection methods, particularly those based on optical detection. Moreover, with the ability to perform simultaneous and identical functions in parallel layouts, immunoassays on a CD are an ideal platform for high throughput screening of analytes. Figure 4 gives a schematic illustration of a CD-based sandwich enzyme-linked immunosorbent assay (ELISA). The key concept is to overcome the capillary force, which prevents the liquid in one chamber from moving out, by using the centrifugal force controlled by the rotational speed of the CD. Furthermore, the increasing centrifugal force from the center toward the edge of the CD is used for flow sequencing which replaces the stepwise procedures in a conventional ELISA. Not only can ELISA be performed on this CD-based platform but also other immunoassays can be realized using a very similar approach. For more detailed information such as the operating theory and instrumentation, readers can refer to various review articles.^43,44

Figure 4.

ELISA in a lab-on-a-CD. (A) The 4-array CD-ELISA design, and (B) a single assay. Chambers 2, 4, 6, and 8 are loaded with washing buffer, whereas chambers W, D, 1, 3, 5, 7, 9 are designed for waste, detection, first antibody, blocking solution, sample, second antibody, and substrate, respectively. As the CD rotates, the solutions in chambers 1 to 9 flow out sequentially because of the decreasing centrifugal force and then complete the ELISA procedures. The most common valves are hydrophobic and capillary valves. Hydrophobic valves stop the flow by the hydrophobic surfaces created in the microchannel, and capillary valves stop the flow by a capillary pressure barrier at junctions where the channel diameter suddenly expands.

This section will focus on some representative works on CD-based immunoassays published between the years 2005 and 2009. The first centrifugal fluidic platform was developed in 1969⁴⁵ and has been extensively studied since then. Companies, such as Gyros AB, Tecan, and Burstein Technologies also developed commercial products. Recently, Gyros AB developed a CD-based immunoassay for the determination of α-fetoprotein (AFP), IL-6, and carcinoembryonic antigen (CEA). The disc structure is an intricate web of microchannels and reaction chambers packed with streptavidin-coated PS microparticles for conjugating with biotinylated Abs against AFP, IL-6, and CEA. Alexa 647-labeled bovine serum albumin (BSA) was used as a detection Ab. All the flow control is through the rotation of the CD mounted on a workstation equipped with a detection system (laser-induced fluorescence, LIF) and robotic arms for dispensing liquids into the inlets on the CD. The experimental results showed that the detection limits for AFP, IL-6, and CEA were 0.15, 1.25, and 1.31 pmol/L, respectively.⁴⁶ The inter-and intra-assay coefficients of variation were less than 10% and 20%, respectively. This disc also demonstrated an excellent analytical efficiency (104 assays in 50 min) when compared with the performance of traditional 96-well ELISA plates (96 assays, a few hours). Moreover, the relationship between the fluorescent imaging profiles of the column structure and the kinetic properties of the Ag and Ab was also investigated.

Although the disc developed by Gyros AB gives an excellent analytical efficiency within acceptable variations, an expensive robotic unit is inevitably required. For practical applications, a cheap and fully automated CD immunoassay without a robotic system is required. To achieve this, valves capable of precisely controlling the flow sequences in an immunoassay are required. For instance, a disc with superhydrophobic capillary valves was reported to perform the sequential flow steps.⁴⁷ A capillary valve is the most popular valving component in CD immunoassays because it requires no moving parts or external actuation. Unfortunately, proteins present in solution may adsorb onto conventional capillary valves to gradually make them hydrophilic, leading to the failure of the valve. To solve this problem, polyaniline nanofibers along with a noncrystalline fluorine coating to create superhydrophobic nanostructures as valves have been reported.⁴⁷ Measurements showed that the contact angle before and after protein blocking were 170° and 165°, respectively, indicating that these valves maintain their superhydrophobic properties after treatment with protein solutions. This excellent result is because of the increased surface roughness and decreased surface energy attributed to the coating of polyaniline nanofibers and noncrystalline fluorine, respectively. Furthermore, to improve the detection sensitivity of the assay, the detection surface was sequentially treated by an oxygen plasma, polyethyleneimine (PEI), tyrosinase, and protein A to increase its binding capacity based on orientation-controlled immobilization. This disc integrates functions, including pumping, valving, fluid splitting, washing, and mixing. Automation, without using a robotic unit, was achieved because the primary Ab and blocking protein were preimmobilized onto the detection area. The substrate, conjugate, washing, secondary Ab, and Ag solution were loaded into corresponding wells for flow sequencing. Comparing with a conventional 96-well ELISA plate, this disc has advantages that included minimized sample/regent consumption, shorter assay time, and a broader dynamic range.

Another strategy to control the transport of liquid in a CD immunoassay is to use active valves. For example, a disc with a laser irradiated ferrowax microvalve (LIFM) that can be actively controlled to open and close by laser irradiation was reported.⁴⁸ LIFMs are valves made of nano-composite materials in which nanosized iron oxide particles are dispersed in paraffin wax. They are phase change-based microvalves and can be fabricated as normally open, normally closed, and reversible microvalves. Low-intensity laser light can melt the paraffin wax with the embedded iron oxide nanoparticles, whereas laser light with a high intensity cannot melt the wax. Using this mechanism, the LIFMs can be actively opened or closed by the irradiation of laser light during the assay process. This disc contains plasma separation channels and chambers for a substrate solution, washing buffer, stopping buffer, mixing, waste, and detection. These chambers are connected by microchannels where LIFMs are fabricated in a normally open or normally closed state. Figure 5 shows a schematic illustration of the disc design and the reaction steps involving in the active opening and closing of the LIFMs. In this disc, valves 3, 6, 9, and 12 are the normally open LIFMs where all other valves are the normally closed LIFMs. Using these LIFMs together with precoated PS microbeads, the concentrations of the antibody to the hepatitis B surface antigen (anti-HBs) and hepatitis B virus surface antigen (HBsAg) from whole blood samples can be determined in an automated manner within 30 min without using robotic arms, which is a much shorter time when compared with conventional ELISA plates (longer than 90 and 120 min for anti-HBs and HBsAg, respectively).

Figure 5.

(A) Schematic illustration of the disk design. (B—G) The reaction principle of the CD immunoassay to detect anti-HBs. The lightning symbols indicate the laser irradiation on the LIFMs. Reprinted from Ref. 48 with permission from the Royal Society of Chemistry (RSC) (Copyright 2009 The Royal Society of Chemistry).

Power-Free Transport

Most of the liquid transport devices require external power sources to drive the liquids. For example, the electro-wetting method requires the application of a voltage, the micropneumatic pump needs compressed air generated by a bulky pump to deflect the membranes, and the centrifugal force-based transport method uses a rotary motor to spin the CD. Taking into consideration the demands of POC applications, a power-free liquid transport method, such as that used in commercial immunoassay strips, is also a promising alternative. For example, Hosokawa et al.⁴⁹ developed a power-free injection method for a microfluidic immunoassay by air evacuation. As shown in Figure 6, a microchip composed of PDMS microchannels and a glass bottom plate was first degassed in a vacuum system to store the pumping energy, followed by covering them with a piece of adhesive tape. When this microchip was brought back to atmospheric conditions, air dissolved into the PDMS flows again through the walls of the microchannels and reservoirs. Once an aliquot was pipetted at the inlet of the microchannel, the dissolution of air reduces air pressure in the microchannelreservoir system, and hence pumps the solution into the microchannel. Because the inlet of the microchannel is narrow and wettable, a capillary retention valve is naturally formed to prevent the microchannel from drying out. When a second aliquot was pipetted at the inlet of the microchannel, the pumping effect occurred again. Using this method, sample and reagents required for performing an immunoassay were sequentially injected into the microchannels for immobilization, blocking, washing, reaction, and detection. It has been used to demonstrate the detection of rabbit IgG and human C-reactive protein (CRP) with a detection limit of 0.21 and 0.42 nM, respectively, which can be performed within 20 min. Power-free liquid transportation is promising because one of the ultimate goals for microfluidic immunoassays is to perform these chips for POC applications outside the laboratory with minimum support equipment.

Figure 6.

Schematic illustration of a microchip based on power-free liquid transport. (A) The chip design, and (B) the pumping mechanism. Reprinted from Ref. 49 with permission from the Royal Society of Chemistry (RSC) (Copyright 2006 The Royal Society of Chemistry).

Heterogeneous Microfluidic Immunoassays

Immobilization is a key step in heterogeneous immunoassays because it greatly influences the specificity and sensitivity. Briefly, the surface of these microfluidic chips has to be treated so that proteins can be immobilized for further detection. In general, heterogeneous microfluidic immunoassays can be classified as surface-based and bead-based immobilization methods. In heterogeneous microfluidic immunoassays, the Abs or Ags are usually immobilized on the surfaces of the microchannels or microbeads. Immobilization on surfaces of microchannels requires additional steps during the microfabrication process and may suffer from poor reproducibility and reliability. In contrast, microbead-based immobilization can be performed outside the microchip and offers a significantly large surface for immobilization, and hence improved performance of the immunoassays.^52,54

Surface-Based Immobilization

Many surface-based immobilization strategies have been developed for microfluidic immunoassays. Different substrate materials, such as PDMS,^55–59 cycloolefin-copolymer,⁶⁰ PS,^61,62 silicon nitride (SN),⁶³ and PMMA,⁶⁴ have been reported for use in microfluidic devices. Among them, surface immobilization on PDMS will be emphasized in this section because it is now the most popular material for fabricating microfluidic devices. For example, a method called“macro-molecules to PDMS transfer” was reported to directly entrap macromolecules for capturing target analytes during the PDMS polymerization step.⁵⁵ In this work, rabbit IgG or anti-CRP solutions were spotted on a Teflon master by a piezoelectric arrayer, followed by pouring liquid PDMS on top. The PDMS membrane was then peeled off after curing, and the molecules (rabbit IgG or anti-CRP) were thus transferred to the PDMS membrane. The rabbit IgG and anti-CRP-conjugated PDMS membranes were then used to capture rheumatoid factor (RF) and free CRP in human serum samples, respectively, followed by conjugating with peroxidaselabeled detection Abs for chemiluminescence detection. Using this method, the sensitivities for CRP and RF measurements were 12.5μg/L and 5.3 IU/mL, respectively. Due to its simplicity and efficiency for immobilization, this method has a great potential to be applied to fabricate PDMS-based microchip for immunoassays.

Alternatively, Hashimoto et al.⁵⁶ developed an electrochemical method to perform localized immobilization of proteins within the channels of a microchip. In this work, a layer of PDMS containing micropost structures was first attached to the inner surface of an upper glass substrate by an oxygen plasma treatment before assembling with a bottom glass substrate where Pt electrodes were deposited. After assembly, a solution containing PEI and heparin was introduced to electrostatically form a PEI/heparin multilayer on the wall of the channel. To obtain an area for protein immobilization, the channel was filled with a solution of potassium bromide, followed by applying an electrical pulse to the patterned electrodes to locally produce Br₂ and HBrO. The oxidation action of HBrO then caused the PEI/heparin multilayer to locally detach from the substrate, and the exposed area (PDMS) was thus ready for localized immobilization because heparin is resistant to the adhesion of many proteins. Using this method, antiglutathione peroxidase (GPX) was locally immobilized to capture GPX, followed by conjugation with Cy3-labeled anti-GPX for fluorescent detection. This breakthrough work provides a simple method for immobilizing proteins in predefined areas and is believed to be promising for surface immobilization.

Supported bilayer membranes (SBMs), such as phospholipid bilayer membranes are biomimetic membranes capable of providing long-term hydrophilic and protein-resistant properties. Therefore, they have been reported for surface immobilization. For example, Phillips and Cheng⁵⁷ used phosphatidylcholine membranes assembled on PDMS as a supported membrane for protein immobilization. In this work, a phosphatidylcholine membrane was first assembled on an oxygen plasma-treated PDMS surface, followed by vesicle fusion to form cell surface ligand GM1-integrated SBMs for capturing the cholera toxin (CT). Then, rabbit anti-CT and a dye-labeled goat antirabbit Ab were sequentially conjugated for fluorescent detection. Using this method, the entire assay can be completed within 25 min with a detection limit of 8 fmol for the measurement of CT. The SBMs, however, may peel away from the solid support when exposed to air. Therefore, reinforced SBMs (r-SBMs) were developed to solve this problem and sensing of Staphylococcus enterotoxin B (SEB) was demonstrated.⁵⁸ In this work, the supported bilayer was prepared by fusing vesicles containing biotin phosphoethanolamine and phosphatidylcholine to the channels, followed by incubation with a streptavidin solution to form the r-SBMs. The biotinlyated rabbit anti-SEB was then introduced to be immobilized on the streptavidin-conjugated SBMs (r-SBMs) to capture SEB, followed by attaching sheep anti-SEB and a dye-labeled detection Ab for fluorescent detection. Using this method, a detection limit of 0.5 ng/mL was obtained. More importantly, the r-SBMs-modified solid support is air stable and is expected to have the capability for long-term storage in a dehydrated format for disposable applications.

Another method for immobilization is to use dextran, which is a biocompatible polymer. The immobilization of dextran on a microchip surface by introducing linkers for protein immobilization is a less toxic and inexpensive process. More importantly, the hydrophilicity and immobilization efficiency can be enhanced because of the abundant oxidation-resultant aldehyde groups. For instance, Yu et al.⁵⁹ used this dextran-based immobilization to simultaneously detect three biomarkers. In this work, a solution containing H₂O and HCl was first introduced to oxidize the surface of the PDMS channels, followed by the introduction of a solution containing 3-aminopropyltriethoxylsilane (APTES) and ethanol to make the surface abundant with amine groups. Then, the partially oxidized dextran (with aldehyde groups) solution was introduced to conjugate onto the channel surface via the formation of shiff's base between the amines and aldehydes, followed by treating with NaIO₄ to oxidize the dextrans again. The surface is now aldehyde abundant and is ready to react with the amine groups of proteins. Capture Abs, such as anti-IL5, anti-HBsAg, and anti-rabbit IgG were immobilized in different channels to capture IL5, HBsAg, and rabbit IgG, respectively, followed by introducing HRP-labeled detection Abs for colorimetric detection. Using this method, a detection limit of 100 pg/mL was obtained for these three markers.

In addition to PDMS, dextran can be also used in a cyclo-olefin-copolymer substrate for protein immobilization. For instance, önsson et al.⁶⁰ first treated a copolymer chip with an oxygen plasma and APTES to expose amine groups, followed by a sequentially treatment with partially oxidized dextran and NaIO₄ to make the channel surface aldehyde abundant. Then, capture Abs for CRP were immobilized for performing immunoassays. Using Alexa 647-labeled detection Abs for fluorescent detection, a detection limit of 2.6 ng/mL was obtained. These dextran-based immobilizations are long-term stable, inexpensive, and capable of being produced in large quantities, resulting in the possibility for commercialization.

Other materials, such as PS, SN, and PMMA have also been used as substrates for microfluidic immunoassays. For example, Darain et al.^61,62 used different strategies to immobilize proteins on PS substrates. In one work,⁶¹ the PS substrate was sequentially treated with ultraviolet (UV)/O₃ (to introduce OH group), APTES (to introduce NH₂ group), and glutaraldehyde (to introduce CHO group), followed by conjugating with antimyoglobin for capturing myoglobin. In another work,⁶² the PS substrate was sequentially treated with sputtered gold (to introduce a Au surface), 16-mercapto-hexadeconic acid (to introduce COOH groups), and N-hydroxy succinimide ester and 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (to introduce active NSH esters), followed by conjugating with antihorse IgG, for capturing horse IgG. After fluorescent detection, the detection limit for myoglobin and horse IgG was obtained as 16 ng/mL and 0.71μg/mL, respectively. These two PS-based immobilization strategies can be extended for the measurement of other important biomarkers for diagnostic applications.

SN has been widely used in the semiconductor industry because of its excellent chemical, electrical, and optical properties. It can also be used as a substrate for microfluidic immunoassays. For instance, He et al.⁶³ used SN as the substrate for immunosensing of CRP in two formats. The first format is a two-layer system in which anti-CRP was directly patterned on a SN-coated silicon wafer as the first layer to capture dye-labeled CRP; the second format is a four-layer system in which biotinylated BSA was first adsorbed on a SN-coated silicon wafer, followed by sequential conjugation with avidin, biotinylated anti-CRP, and dye-labeled CRP. The biotinylated BSA works as the first adhesion layer and as a blocking material as well. Therefore, another blocking solution is not required in this format. Using 1.76μM of CRP as the analyte, the fluorescence intensity obtained from the four-layer system is 3.2-fold stronger than that obtained from a two-layer system.

PMMA has been used as a substrate for microfluidic devices because of its relatively low cost and optical transparency. For example, Wang et al.⁶⁴ used PMMA microchannels filled with an Al₂O₃ sol—gel network to encapsulate capture Abs for immunoassays. In this work, the inner surface of the PMMA microchannel was first treated with a copolymer to bear silanol groups, followed by introducing a solution containing anti-IgG and Al₂O₃ sol gel to allow polycondensation between the sol gel and the silanol groups. The anti-IgG was thus encapsulated in the network for capturing target IgG. After conjugating with alkaline phosphatase (ALP)-labeled secondary Abs, electrochemical detection was used with a detection limit of 1 pg/mL. The excellent sensitivity is attributed to the high surface area-to-volume ratio of the network. Table 3 summarizes the aforementioned surface-based immobilization strategies. The substrate selection, target analyte, immobilization scheme, detection strategy, and detection limits have all been summarized.

Table 3.

Summary of surface immobilization strategies in microfluidic immunoassays

				Detection strategy
Ref.	Substrate	Target analyte	Immobilization scheme	(detection limit)
55	PDMS	1) RF	PDMS/RF; CRP (direct entrapment)	Chemiluminescent,
		2) CRP		1) 5.3 IU/mL
				2) 12.5 μg/L
56	PDMS	GPX	PDMS/(PEI + heparin)/anti-GPX	Fluorescent, 5 ng mL
57	PDMS	CT	PDMS/SBMs/GMI	Fluorescent, 8 fmol
58	PDMS	SEB	PDMS/rSBMs/biotinlyated rabbit anti-SEB	Fluorescent, 0.5 ng/mL
59	PDMS	1) IL5	PDMS/Dextran/anti-IL5; anti-HBsAg;	Colorimetric,
		2) HBsAg	anti-rabbit IgG	I) 100 pg/mL
		3) rabbit IgG,		2) 100 pg/mL
				3) 100 pg/mL
60	Cycloolefin-copolymer	CRP	cycloolefin-copolymer/dextran/anti-CRP	Fluorescent, 2.6 ng/mL
61	PS	Myoglobin	PS/UV+O₃/APTES/glutaraldehyde/anti-myoglobin	Fluorescent, 16 ng/mL
62	PS	Horse IgG	PS/Au/16-MHA/NHS ester+EDC/anti-horse IgG	Fluorescent, 0.71 μg/mL
63	SN	CRP	1) SN/anti-CRP	Fluorescent, less than 1.76 μM
			2) SN/biotinylated BSA/avidin/biotinylated anti-CRP
64	PMMA	IgG	PMMS/AI₂O₃ sol—gel/anti-IgG (encapsulate)	Electrochemical, I pg/mL

Bead-Based Immobilization

As mentioned previously, microbead-based immobilization offers a large surface area and hence may improve the performance of the immunoassays. In the following sections, bead-based immobilization methods classified as either nonmagnetic or magnetic microbeads in microfluidic immunoassays will be reviewed and discussed.

Nonmagnetic Microbeads. The most commonly used nonmagnetic microbeads in microfluidic heterogeneous immunoassays are PS beads and glass beads because of their extensively studied characteristics. For example, Ko et al.⁶⁵ developed a microfluidic electroimmunosensing chip for simultaneous determination of cancer biomarkers, including AFP, CEA, and prostate-specific antigen by a combination of Ab-conjugated PS beads (for immobilization), gold nanoparticles (for labeling), and silver enhancer (for signal amplification). In this work, PS beads were coated with strepavidin for conjugation with biotinylated capture Abs. These capture Ab-conjugated beads were then introduced and trapped in the reaction area by PDMS micropillars, followed by introducing samples containing cancer markers, gold nanoparticles-labeled detection Ab, and silver enhancer, respectively. Gold nanoparticles were used to catalyze the reduction of silver ions into metallic silver, which was then deposited onto the gold nanoparticles and enlarged the particles to 30–100 nm. Finally, an electrical bridge was formed between the underlayered Pt electrodes and increased the electrical path, resulting in a decrease in resistance (Fig. 7). Compared with conventional immunoassays, the analysis time was reduced from several hours to 55 min.

Figure 7.

Alternatively, Lee et al.⁶⁶ used glass beads to detect a bacterial pathogen. In this work, glass beads were sequentially treated with 3-aminopropyl trimethoxysilane, glutaraldehyde, and a capture Ab (anti-O157 Ab). These beads were then packed in spatially isolated microchambers, followed by introducing l-lysine solution for blocking unreacted surfaces. Samples containing E. coli O157:H7 expressing green fluorescence protein were then introduced and followed by fluorescence detection.

Most of the previous works only used the outer surface of the beads for immobilization. To improve the sensitivity of an immunoassay, Yang et al.⁶⁷ used superporous agarose beads (superpores with diameters ranging from 10 to 80 μm) as a solid support for performing microfluidic immunoassays. In this work, superporous agarose beads were sequentially treated with protein A and anti-goat IgG to capture target goat IgG, followed by treatment with ALP-conjugated anti-goat IgG and 5-bromo-4-chloro-3-indolyphosphate/nitroblue terazolium for colorimetric detection. Using these superporous beads as immobilization materials, a detection limit of 100 pg/mL of goat IgG was obtained, which is 10 times more sensitive than conventional bead-based microfluidic immunoassays.⁶⁸ The improved sensitivity is mainly because of the large surface area attributed to the inner matrices of the super-porous beads.

Most nonmagnetic bead-based microfluidic immunoassays require the fabrication of microstructures to retain the beads for immunoreactions. Alternatively, Sivagnanam et al.⁶⁹ used a brilliant strategy to retain the beads. In this work, positively charged APTES were patterned on a glass substrate, followed by the electrostatic assembly of negatively charged streptavidin-coated microbeads. These self-assembled streptavidin-coated microbeads were then used to conjugate with a biotinylated capture Ab to capture the target Ags. Using this method, a target mouse IgG Ag was detected with a detection limit of 15 ng/mL and 250 pg/mL in a stop-flow and a continuous-flow mode, respectively. This approach requires no microstructures to retain beads and is promising for patterning complicated immobilization beads for multiplexed applications.

Bead-based immunoassays have also been used in flow cytometry since the 1980s,⁷⁰ with an improved sensitivity, specificity, and dynamic range.^71,72 Recently, Holmes et al.⁵³ developed a bead-based immunoassay performed on a microchip-based flow cytometer. In this work, commercial glycidyl methacrylate microbeads were sequentially treated with succinic anhydride, protein A, and human IgG or rabbit IgG to form human IgG or rabbit IgG-conjugated microbeads, respectively. These microbeads were then reacted with samples containing Cy3-labeled anti-human IgG and Cy5-labeled anti-rabbit IgG to form the complexes, followed by analysis in a micro flow cytometer. The micro flow cytometer used negative dielectrophoresis to focus the microbeads into a particle stream, and the impedance and fluorescence signals were measured for indicating particle size and for obtaining scatter plots, respectively. The results were comparable to those obtained by a commercial flow cytometer. Although the data from measurements of particle sizes were not used (accurate sizing of beads with diameters ranging from 2 to 10 μm), it has a great potential for bead encoding. Combining the bead encoding with fluorescence detection in a micro flow cytometer is believed to be a promising platform for multiplexed diagnosis.

Magnetic Beads. Magnetic beads are popular for a fully automated detection system because they have a large surface area, are easy to manipulate by a magnetic field, and allow easy separation of target analytes from the reaction mixture. In contrast to nonmagnetic beads that usually require microstructures to retain the beads, magnetic beads can be easily retained at desired areas by the application of a magnetic field.^{73 –75} For example, Lien et al.⁷⁶ developed an integrated microfluidic system for virus detection by using Ab-conjugated magnetic beads to capture target viruses. In this work, the target viruses (dengue virus serotype 2 and enterovirus 71) were first captured by the Ab-conjugated magnetic beads for enrichment under a magnetic field. Then, the virus RNA was extracted and transcripted to complementary deoxyribonucleic acid, followed by an amplification process using a micro-RT-PCR module. The required components for the entire assay, such as micropumps/microvalves for liquid delivery, microcoils for magnetic field generation, and microheaters/microsensors for temperature control were all integrated on a single chip. Using this highly integrated microdevice, viruses were successfully purified, collected, and detected in an automatic fashion with a detection limit of 10² pfu/mL. Compared with large-scale equipment, this integrated microsystem can automatically perform mixing, incubation, purification, transportation, and nucleic acid amplification of a virus, making it a flexible platform for future diagnostic applications. Magnetic bead-based microfluidic chip for the separation and detection of bacteria was also reported. For example, Qiua et al.⁷⁷ used a U-type microchip in which anti-Staphylococcus aureus (S. aureus) and anti-Salmonella-modified magnetic beads were magnetically immobilized to specifically capture S. aureus and Salmonella, followed by fluorescent detection through fluorescein isothiocyanate-labeled anti-S. aureus and anti-Salmonella, respectively.

There are a lot of studies on magnetic bead-based microfluidic immunoassays. Readers can refer to several articles for different applications, such as the determination of AFP,⁷⁸ cholera toxin subunit B,⁷⁹ and multiplexed immunoassays.⁸⁰

Homogeneous Microfluidic Immunoassays

Unlike heterogeneous immunoassays performed in microfluidic devices, a concentration step is usually required in microchip-based homogeneous immunoassays because the surface of the microchannel is not used for immobilization. Homogeneous microfluidic immunoassays are mostly performed in an electrophoresis-based microchip format, and many concentration strategies, such as the use of a preconcentration membrane or isotachophoresis (ITP) to stack the target analytes have been used for improving the sensitivity of detection. For instance, Meagher et al.⁸¹ developed an electrophoretic immunoassay-based microchip incorporated with a photopolymerized membrane for preconcentration and polyacrylamide gel-filled microchannels for sample/reagent loading and Ab/complex separation. As shown in Figure 8, protein biotoxins (sample) were loaded through a larger pore-size polyacrylamide gel to upstream of the membrane, followed by trapping for preconcentration under the continuous application of an electric field. The capability of the membrane to perform preconcentration arises from its porous structure that allows the transport of buffer ions but excludes the proteins. Then, dye-labeled Abs were loaded and concentrated in the same area. After mixing and conjugating, unbound dye-labeled Abs and complexes were separated in the separation channel in which a smaller pore-size polyacrylamide gel was filled on switching the direction of the electric field. Using this method, the entire immunoassay can be performed within 20 min with a minimal sample volume of less than 10 μL and a detection limit of less than 10 pM for SEB determination. This microchip was also used for the determination of collagen-cleaving enzyme matrix metalloproteinase-8 for rapid saliva-based clinical diagnostics.⁸² Although the aforementioned microchip performs very well for preconcentration and analysis of target analytes, a microchannel for the separation of unbound labeled Abs and the complexes was required. In another work, Reichmuth et al.⁸³ fabricated a membrane with proper pores (about 10 nm) in the microchannel. Due to these larger pores, this membrane allows the transport of both buffer ions and unbound Abs but excludes the viral particles (about 80 nm). Using this membrane-based concentration and filtration method, rapid detection of swine influenza virus was successfully performed within 6 min with a detection limit of 610 Truckee—Carson Irrigation District (TCID) 50/mL without an electrophoretic separation.

Figure 8.

Schematic illustration of the electroosmotic microchip incorporated with a preconcentration membrane and a separation channel. S = sample, SW = sample waste, Ab = antibody, B = buffer (not used here), LB = loading buffer, RB = running buffer, RW = running waste. Reprinted from Ref. 81 with permission from the Royal Society of Chemistry (RSC) (Copyright 2008 The Royal Society of Chemistry).

ITP⁸⁴ is a commonly used preconcentration method in capillary electrophoresis and microchip electrophoresis. For instance, Kawabata et al.⁸⁵ developed an“electrokinetic analyte transport assay” microchip in which ITP was performed for on-chip concentration, mixing, and reaction, followed by electrophoretic separation and detection in the downstream channel. As shown in Figure 9, an electrophoretic microchip was designed for the formation of plugs of the tailing buffer, the first Ab, the sample, the secondary Ab, and the leading buffer in the channel by the application of vacuum pressure at the waste wells. On the application of an electric field between reservoirs 1 and 6, the first Ab was concentrated by ITP and migrated into the sample plug to form the first Ab—sample complex. This complex then migrated into a dye labeled-secondary Ab plug to form the sandwich complex (first Ab—sample-labeled secondary Ab). This sandwich complex was then separated from the other components by using a microchip gel electrophoresis through the application of an electric field between reservoirs 5 and 6. Using this ITP-integrated electrophoretic microfluidic immunoassay chip, AFP was successfully detected within 136 s with a detection limit of 5 pM. Similarly, Kagebayashi et al.⁸⁶ also used a microchip together with a desktop instrument containing components functionalized for chip loading/transportation/storage, pipetting, vacuum applications, conductivity monitoring, electric field generation, sample and reagent storage, and LIF detection to determine total AFP from serum samples. Good correlation with a commercially available reference method (LiBASys) for 68 patient serum samples was obtained. In summary, these on-chip electrophoretic immunoassays using membranes or ITP-based concentration and gel separation have a great potential to provide a rapid, sensitive, portable, and cost-effective method for both clinical and research applications.

Figure 9.

Schematic illustration of the ITP-integrated microfluidic immunoassay chip. W = waste, 1 = tailing buffer, 2 = first antibody (DNA coupled anti-AFP WA1 antibody), 3 = sample (AFP), 4 = second antibody (Hilyte-labeled anti-AFP WA2 antibody), 5 = handoff well, 6 = leading buffer. Reprinted from Ref. 85 with permission from Wiley InterScience (Copyright 2008 Wiley-VCH Verlag GmbH & Co. KGaA)

Detection in Microfluidic Immunoassays

Generally speaking, in an immunoassay, one of the immunoagents must be conjugated with specific labeling molecules to facilitate the detection of the analytes. The most common detection techniques used in microfluidic immunoassays are fluorescent^{41,46,47,49,53,56 –58,60 –63,65,69,77,81 –83,86} and lumines-cencent^37,55 detections because of their excellent sensitivity. Electrochemical detection is also popular because of its capability for miniaturization and simple instrumentation.^64,65 Readers can refer to the recommended references for more information about these detection methods. In the following sections, recent works in microfluidic immunoassays using different detection strategies, except those previously aforementioned detection schemes, will be reviewed and discussed. We mainly focus on label-free detection methods, including surface plasmon resonance (SPR), a thermal lens microscope (TLM), quartz crystal microbalance (QCM), and wavelength-interrogated optical sensor (WIOS) schemes. Detection methods based on the labeling of gold nanoparticles (colorimetric detection) and magnetic microbeads (magnetoelectric detection) are also introduced. These methods are classified as optical (SPR, TLM, WIOS, colorimetric detection) and nonoptical (QCM, magnetoelectronic detection) detections in the following sections.

Optical Detection

Surface Plasmon Resonance. SPR is a powerful detection strategy for immunoassays because of its capability to perform real-time sensing of interactions between Ab and Ag without the use of labeling molecules. Information, such as the concentration of analytes, kinetics, and affinity can be readily obtained. Recently, Lee et al.⁸⁷ developed a microfluidic system integrated with 2-D SPR phase imaging systems for microarray immunoassays (Fig. 10). In this work, the microfluidic device was composed of an arrayed sample/reagent delivering module for well-controlled and high throughput fluidic transportation and a microtemperature control module for maintaining the temperature field within the arrayed detection area for reducing the thermal noise during SPR sensing. The fluidic transportation module was composed of arrayed microchannels, micropumps, microvalves, and flow sensors, whereas the temperature control module was composed of micromachine-based temperature heaters and sensors fabricated on a soda-lime glass substrate. Using this system, the interaction of anti-rabbit IgG and IgG was successfully detected and a detection limit of 1 × 10⁻⁴ mg/mL for IgG was obtained. Microfluidic devices based on SPR sensing for the determination of disease markers⁸⁸ and environmental toxins⁸⁹ have also been reported. Furthermore, portable, compact, and low-cost SPR instruments, such as the Spreeta 2000 developed by Texas Instruments Inc. are now commercially available. It is believed that combining a microfluidic immunoassay platform with these cheap and miniaturized SPR instruments provides a promising tool for practical biochemical sensing applications.

Figure 10.

(A) Schematic illustration of a microfluidic system integrated with a 2-D SPR phase imaging system. (B) Schematic illustration of the microfluidic device comprising of a fluidic transport module and a temperature control module. Reprinted from Ref. 87 with permission from Elsevier (Copyright 2007 Elsevier).

Thermal Lens Microscope. Fluorescence detection is the most commonly used detection strategy in bioanalysis because of its high sensitivity. However, the protocol for labeling fluorophores is tedious and laborious. Alternatively, TLM is highly suitable for the quantization of an ultra-low trace amount of analytes in small-volume samples, especially for nonfluorescent samples.⁹⁰ In TLM, the coaxial beams of the excitation and the probe beam are focused into the sample stream. The sample molecules are excited by the irradiation of the excitation beam, followed by a decay back to a ground state through the releasing of heat and hence causing localized temperature increases in samples. The refractive index of the localized medium then changes because of the temperature increase and produces an effect to form a lens in the medium, which is named as a“thermal lens.” This thermal lens then causes the divergence of the probe beam. Because the size of the thermal lens linearly depends on the number of excited sample molecules, quantitative analysis can be performed by measuring this probe beam divergence. For example, Henares et al.⁹¹ developed a multiple micro ELISA system to simultaneously determine human, goat, and chicken IgGs by combining a capillary-assembled microchip and TLM. In this work, plugs of square capillaries possessing valving and immunoreaction functions were inserted into a PDMS microchannel network (Fig. 11). The valving and immunoreaction functions were performed by inserted capillaries immobilized by a thermally responsive polymer and anti-IgGs, respectively. Using this multiple micro ELISA system, multiple determinations to a precision of 0.1 ng/mL for IgGs (human, goat, and chicken) can be performed within 30 min. Although only the determination of IgGs was performed in this work, it is expected that this system can be applied to many other immunoapplications because of the flexible design of the multiple micro ELISA chip, which allows for the insertion of any required Ab-immobilized capillaries. Moreover, a miniaturized TLM was developed by making use of laser diode modules and fiber-based optics.⁹² By combining this multiple micro ELISA chip with a miniaturized TLM, it is believed that it will have practical use in the near future.

Figure 11.

Wavelength Interrogated Optical Sensing. WIOS is a recently developed, label-free detection method.⁹³ As shown in Figure 12, a wavelength-tunable light source (a collimated light beam from a vertical cavity surface emitting laser) was used to excite waveguide modes in a dielectric waveguide by an input grating coupler. Then, the light is transmitted to a photodetector through an output grating coupler if any resonance occurs. Because the wavelength is continuously scanned, the output signal represents a resonance peak from the grating coupler in which the wavelength position indicates the amount of the analytes adsorbed. This method has been used for microfluidic immunoassays. For instance, Suarez et al.⁹⁴ developed a WIOS-based microfluidic chip for the simultaneous determination of residual antibiotics in raw milk. In this work, the microfluidic chip was composed of a PMMA reservoir plate, a PMMA holder, and a UV-curable hyperbranched polymer sensing chamber, where specific haptens for antibiotics, including fulfonamides, fluorequinolones, and neutravidin were coated. The whole immunoassay procedure including the introduction of a sample solution (raw milk), washing buffer, signal amplification buffer (secondary Ab), and detection can be performed within 10 min with an accuracy of 95%. In contrast to techniques requiring a laboratory environment, such as high-performance liquid chromatography UV/visible spectroscopy,⁹⁵ high-performance liquid chromatography-mass spectrometry,⁹⁶ ELISA,⁹⁷ and parallel affinity sensor array,⁹⁸ this WIOS-based microchip is a promising platform for daily, fast, and automated determinations of antibiotics outside the laboratory by integrating micropumps and microvalves onto the microchip.

Figure 12.

Colorimetric Detection. Although microfluidic immunoassays are a potential platform for POC applications, there is still a need for an easy-to-use, portable, user-friendly, and cheap detection strategy to make it acceptable for practical or even personal use. Colorimetric detection may fulfill such requirements.⁹⁹ Recently, Lin et al.¹⁰⁰ developed a colorimetric detection strategy by using a nanoparticle-labeled Ab to realize the detection of H. pylori- and E. coli O157:H7-specific Ags based on the light-scattering properties of gold nanoparticles. In this work, lysate Ags of H. pylori were adsorbed onto the surfaces of the microchannels, followed by sequentially introducing biotin-conjugated rabbit polyclonal anti-H. pylori Ab and gold nanoparticles functionalized with goat anti-biotin Ab. For the detection of an E. coli O157:H7-specific Ag, similar steps were performed by replacing the lysate Ags of the H. pylori and biotin-conjugated rabbit polyclonal anti-H. pylori Ab with lysate Ags of E. coli O157:H7 and biotin-conjugated goat polyclonal anti-E. coli O157:H7 Ab, respectively. With the light-scattering properties of gold nanoparticles, the color differences quantified by software custom developed for this study increases with the increased concentrations of H. pylori- and E. coli O157:H7-specific Ags. The detection limits for both Ags are 10 ng, which is comparable to those performed by conventional dot blot assays. Another work by Lu et al.¹⁰¹ used a gold nanoparticles-enhanced silver staining technique to detect human IgG (hIgG). In this work, the microfluidic device was composed of a PS bottom layer and PDMS microchannels. The PS plate was first immobilized with goat anti-human IgG (g-hIgG) for capturing hIgG in analytes. After the introduction and incubation of an analyte solution, gold nanoparticle-labeled g-hIgG was introduced to form the complex. Then, to facilitate the visualization of nanoparticle labels, a signal amplification method was performed in which silver ions were reduced by hydroquinone to silver metal at the surfaces of the gold nanoparticles. Finally, a cell phone with an embedded camera was used to snap photos of the microchannels followed by analysis of these photos in a personal computer with image processing and analysis software. Using this method, hIgG concentrations as low as 1 ng/mL can be determined. This detection scheme offers advantages, including low cost, portability, and ease of use and is promising for POC applications.

Nonoptical Detection

Quartz Crystal Microbalance. The QCM is a mass-sensitive detection method for measuring the resonance frequency shift attributed to mass changes on a surface and works well without using any labeling molecules.¹⁰² Therefore, it is suitable for sensing the binding between Ags and Abs in microfluidic immunoassays. Recently, Michalzik et al.¹⁰³ demonstrated a miniaturized system for the determination of protein A by integrating a QCM sensor and a PDMS-based microfluidic cell. First, the quartz material was chemically etched to form a suitable geometry such that the mass sensitivity can be enhanced, followed by the deposition of gold electrodes. For immunosensing, a self-assembled monolayer (SAM) was fabricated on top of the gold electrodes, followed by immobilizing protein A, BSA, and anti-protein A, sequentially. BSA was used to block the nonoccupied surface on the sensing area. Using this method, measurements of protein A with a concentration of 1 mg/mL can be successfully performed. The reusability of the developed system by regenerating the sensing surface with sodium hydroxide has also been demonstrated. Moreover, with the protein A layer, it is possible to immobilize the corresponding Ab layer on the surface for sensing specific Ags. This system, therefore, provides advantages including online measurement, label-free sensing, and reusability, making it potentially useful for microfluidic immunoassays.

Magnetoelectronic Detection. In most immunoassays, nanoscale labeling molecules, such as fluorophores, chromophores, luminescent species, nanoparticles, enzymes, or electrochemi-cally active species are commonly used to conjugate with target analytes for detection because of their comparable sizes. Recently, Mulvaney et. al.¹⁰⁴ reported that using micrometer scale labels, such as magnetic microbeads offers significant advantages and demonstrated a platform combining microfluidics and magnetoelectronics to determine the concentration of ricin A chain (RCA) in samples. In this work, a standard microscope slide was treated in a series of steps to bear a NeutrAvidin layer followed by spotting biotinylated goat anti-ricin onto it. Samples containing RCA were then introduced for incubation, followed by introducing rabbit anti-RCA for incubation. Then, sheep anti-rabbit Ab-coated magnetic microbeads were introduced. Finally, a controlled laminar flow generating a drag force on each bead that is less than the rupture strength of the immunocomplex but greater than that required to remove nonspecifically bound microbeads was applied at the capture surface. Counting of magnetoelectronic beads was finally performed by using a magnetic sensor by measuring the resistance response. This platform was also applied to the determination of femto-molar DNA concentrations without performing PCR amplification. This work demonstrated a platform capable of simplifying assay procedures but providing excellent sensitivity and specificity for the determination of DNA and proteins in complex matrices. With a combination of portable magnetoelectronic sensors, this platform is expected to be a simple but reliable method for microfluidic immunoassays.

Conclusions and Future Perspectives

Microfluidic immunoassays may play an important role in the areas of clinical diagnostics, pharmaceutical analysis, environmental and food safety surveillance, and basic scientific investigations because of the capabilities for miniaturization, integration, and automation. However, to achieve this objective, integrating knowledge from fields, such as material characterization, fabrication methods, liquid transportation, surface modification, immobilization, and detection schemes need to be optimized and integrated. Therefore, critical issues in the following fields are important and should be addressed properly.

Mass Production

For commercial and disposability considerations, polymeric substrates are the dominant candidates in this field. Although PDMS has been widely used for fabricating prototypes of microfluidic devices for immunoassays, the fabrication process is still limited to a replication method involving several hours of curing time. Materials with similar properties and capabilities for mass production by techniques, such as injection molding and embossing are in great demand for practical applications.

Multiplicity

An important feature of microfluidic immunoassays is the ability to perform a multiplexed assay in a single chip. Recently, Zhao et al.¹⁰⁵ and Sun et al.¹⁰⁶ developed a suspension array for a multiplexed immunoassay using silica colloidal crystal beads (SCCBs) exhibiting different reflection spectra as carriers. After the reaction between a sample solution containing fluorescence dye-labeled anti-hIgG and anti-mouse IgG with a SCCBs suspension containing human IgG or mouse IgG-conjugated SCCBs, a multiplexed immunoassay can be successfully performed by a color-encoded method. By combining microfluidic devices with SCCBs-based multiplex immunoassays, it is believed to be a potential platform for clinical applications. The multiplexed assay will continue to be the dominant method for commercialization of these microfluidic immunoassays.

Surface Modification and Immobilization

Nonspecific adsorption or binding to molecules rather than analytes is a key concern in immunoassays because it may greatly impair the sensitivity and selectivity. This is a challenging issue in microfluidic immunoassays. Fortunately, surface chemistry for the functional modification of commonly used materials including device substrates and microbeads/microspheres has been well studied and developed and is expected to provide a solid foundation in microfluidic immunoassays. However, surface modification and immobilization of these new materials still requires tremendous research efforts.

Purification and Concentration

Due to the complexity and the low trace of analytes in biosamples, purification and concentration steps are often required in microfluidic immunoassays. Currently, micro-or nanoparticles and magnetic beads have been widely used for improving the sensitivity by performing a purification process. With a high surface area and ease of manipulation, these beads are expected to provide a promising method for one-step purification and concentration by combing these two techniques together (e.g., beads-on-beads configuration)¹⁰⁷ in microfluidic immunoassays.

Detection

Detection systems for microfluidic immunoassays are usually bulky and expensive, in contrast to other microcompo-nents, such as micropumps, microvalves, micromixers, and microheaters. Although some works on integrated detection system¹⁰⁸ have been developed, the cost, sensitivity, and fabrication processes still hinder their practical applications. Therefore, developing and integrating miniaturized, compact, portable, and inexpensive detection systems with an acceptable sensitivity onto microfluidic devices are still in great demand.

Integration, Packaging, and Cost down Issues

The ultimate goal for practical commercialization is to develop fully integrated, well packaged, disposable, and cheap microfluidic systems for immunoassays. Although many microfluidic devices have been demonstrated, most of them are operated by well-trained personnel in the laboratories and off-chip peripheral units, such as power sources, controllers, and detection systems are used. Moreover, the device cost is still relatively high because they are not massively produced. These factors greatly hinder their commercial applications. Fortunately, some efforts are made to address these problems in recent years. For example, a small-sized blood analyzer comprising a disc loader, a disc positioning, and rotation controller, a laser position controller, an optical detection unit, a temperature controller, and a user interface controller was developed by Lee et al.⁴⁸ Similarly, a prototype incorporating electronics, optics, and chip architecture for high throughput toxin diagnostics was developed by Meagher et al.⁸¹ A desktop instrument containing the LIF detection system, mechanisms for chip loading and transport, electrophoresis power supply, a means for monitoring the buffer conductivity, chip storage cassettes, and refrigerated reagent storage compartment was developed by Kagebayashi et al.⁸⁶

It is envisioned that an integrated low-cost microfluidic immunoassays with a multiplexed detection function can be realized in the near future.

Storage of Reagents

Although many portable microfluidic devices have been well developed and applied in immunoassays in recent years, the long-term storage of bioreagents in portable microfluidic devices is rarely discussed. This is one important issue in portable devices because most bioreagents are not durable under room temperature and some of them require special environments for storage. Therefore, further efforts are required to address these demands.

Acknowledgment

The authors gratefully acknowledge the financial support provided by the National Science Council in Taiwan (NSC 96-2628-E-006-238-MY3).

Competing Interests Statement: The authors certify that they have no relevant financial interests in this manuscript.

References

Ibrahim

Darwish

I. A.

Al-Obaid

A. R. M.

Al-Malaq

H. A.

New highly sensitive enzyme immunoassay for the determination of pravastatin in human plasma. Talanta 2009, 79(5), 1478–1483.

Minkstimiene

A. K.

Ramanaviciene

Ramanavicius

Surface plasmon resonance biosensor for direct detection of antibodies against human growth hormone. Analyst 2009, 134(10), 2051–2057.

Blazžková

Koets

Rauch

Amerongen

A. V.

Development of a nucleic acid lateral flow immunoassay for simultaneous detection of Listeria spp. and Listeria monocytogenes in food. Eur. Food Res. Technol. 2009, 229(6), 867–874.

Rica

R. D. L.

Baldi

Sánchez

C. F.

Matsui

Single-cell pathogen detection with a reverse-phase immunoassay on impedimetric transducers. Anal. Chem. 2009, 81(18), 7732–7736.

Ramírez

N. B.

Salgado

A. M.

Valdman

The evolution and developments of immunosensors for health and environmental monitoring: Problems and perspectives. Braz. J. Chem. Eng. 2009, 26(2), 227–249.

Kindt

T. J.

Osborne

B. A.

Goldsby

R. A.

Kuby Immunology. 6th ed. New York, NY: W.H. Freeman and Company; 2007.

Gosling

J. P.

Immunoassays: A Practical Approach. Oxford, UK: University Press; 2000.

Wild

The Immunoassay Handbook. 3rd ed. Kidlington, UK: Elsevier Ltd; 2005.

Kohlheyer

Eijkel

J. C. T.

Schlautmann

Berg

A. V. D.

Schasfoort

R. B. M.

Bubble-free operation of a microfluidic free-flow electrophoresis chip with integrated Pt electrodes. Anal. Chem. 2008, 80(11), 4111–4118.

10.

K. W.

Park

Namkoong

Kim

Ock

K. S.

Kim

Y. A.

Cho

Y. K.

World-to-chip microfluidic interface with built-in valves for multichamber chip-based PCR assays. Lab Chip 2005, 5(8), 845–850.

11.

Reyes

D. R.

Iossifidis

Auroux

P. A.

Manz

Micro total analysis systems. 1. Introduction, theory, and technology. Anal. Chem. 2002, 74(12), 2623–2636.

12.

Auroux

P. A.

Iossifidis

Reyes

D. R.

Manz

Micro total analysis systems. 2. Analytical standard operations and applications. Anal. Chem. 2002, 74(12), 2637–2652.

13.

Vilkner

Janasek

Manz

Micro total analysis systems. Recent developments. Anal. Chem. 2004, 76(12), 3373–3386.

14.

Lee

S. J.

Lee

S. Y.

Micro total analysis system (micro-TAS) in biotechnology. Appl. Microbiol. Biotechnol. 2004, 64(3), 288–299.

15.

van den Berg

Lammerink

T. S. J.

Micro total analysis systems: Microfluidic aspects, integration concept and applications In Microsystem Technology in Chemistry and Life Science; van den Berg

Lammerink

T. S. J.

, Eds.; Springer Verlag: Heidelberg, Berlin; 1997.

16.

Rossier

J. S.

Girault

H. H.

Enzyme linked immunosorbent assay on a microchip with electrochemical detection. Lab Chip 2001, 1(2), 153–157.

17.

Pal

Sato

Various shapes of silicon freestanding microfluidic channels and microstructures in one-step lithography. J. Micromech. Microeng. 2009, 19(5), 055003–055013.

18.

Sökmen

Stranz

Fündling

Wehmann

H. H.

Bandalo

Bora

Tornow

Waag

Peiner

Capabilities of ICP-RIE cryogenic dry etching of silicon: Review of exemplary microstructures. J. Micromech. Microeng. 2009, 19(10), 105005–105012.

19.

Rodriguez

Spicar-Mihalic

Kuyper

C. L.

Fiorini

G. S.

Chiu

D. T.

Rapid prototyping of glass microchannels. Anal. Chim. Acta 2003, 496(1–2), 205–215.

20.

Park

J. H.

Lee

N. E.

Lee

Park

J. S.

Park

H. D.

Deep dry etching of borosilicate glass using SF₆ and SF₆/Ar inductively coupled plasmas. Microelectron. Eng. 2005, 82(2), 119–128.

21.

Young

K. C.

Lien

H. M.

Lin

C. C.

Chang

T. T.

Lee

G. B.

Chen

S. H.

Microchip and capillary electrophoresis for quantitative analysis of hepatitis C virus based on RT-competitive PCR. Talanta 2002, 56(2), 323–330.

22.

Wang

Chen

Soper

S. A.

A high-performance polycarbonate electrophoresis microchip with integrated three-electrode system for end-channel amperometric detection. Electrophoresis 2008, 29(9), 1881–1888.

23.

Prest

J. E.

Fielden

P. R.

Goddard

N. J.

Brown

B. J. T.

Isotachophoretic analysis using injection-moulded polystyrene chip devices. Meas. Sci. Technol. 2008, 19(6), 065801–065807.

24.

Lin

Y. H.

Lee

C. H.

Lee

G. B.

Droplet formation utilizing controllable moving-wall structures for double-emulsion applications. J. Microelectromech. Syst. 2008, 17(3), 573–581.

25.

McDonald

J. C.

Duffy

D. C.

Anderson

J. R.

Chiu

D. T.

Schueller

O. J. A.

Whitesides

G. M.

Fabrication of microfluidic systems in poly(dimethylsiloxane). Electrophoresis 2000, 21(1), 27–40.

26.

Abbasi

Mirzadeh

Katbab

A. A.

Bulk and surface modification of silicone rubber for biomedical applications. Polym. Int. 2002, 51(10), 882–888.

27.

Xiao

X. D.

J. J.

Chen

H. Y.

Bulk modification of PDMS microchips by an amphiphilic copolymer. Electrophoresis 2007, 28(18), 3302–3307.

28.

Sung

W. C.

Chang

C. C.

Makamba

Chen

S. H.

Long-term affinity modification on poly(dimethylsiloxane) substrate and its application for ELISA analysis. Anal. Chem. 2008, 80(5), 1529–1535.

29.

Makamba

Hsieh

Y. Y.

Sung

W. C.

Chen

S. H.

Stable permanently hydrophilic protein-resistant thin-film coatings on poly(dimethylsiloxane) substrates by electrostatic self-assembly and chemical cross-linking. Anal. Chem. 2005, 77(13), 3971–3978.

30.

Wong

C. M.

Surface molecular property modifications for poly(dimethylsiloxane) (PDMS) based microfluidic devices. Microfluid. Nanofluidics 2009, 7(3), 291–306.

31.

J. M. K.

Gitlin

Stroock

A. D. S.

Whitesides

G. M.

Components for integrated poly(dimethylsiloxane) microfluidic systems. Electrophoresis 2002, 23(20), 3461–3473.

32.

Wang

C. H.

Chen

Y. Y.

Liao

C. S.

Hsieh

T. M.

Luo

C. H.

J. J.

Lee

H. H.

Lee

G. B.

Circulating polymerase chain reaction chips utilizing multiple-membrane activation. J. Micromech. Microeng. 2007, 17(2), 367–375.

33.

Moğulkoc

Jansen

H. V.

Berenschot

J. W.

Brake

H. J. M.

Knowles

K. M.

Elwenspoek

M. C.

Characterization of MEMS-on-tube assembly: Reflow bonding of borosilicate glass (Duran®) tubes to silicon substrates. J. Micromech. Microeng. 2009, 19(8), 085027–085036.

34.

Hayes

M. A.

Kheterpal

Ewing

A. G.

Effects of buffer pH on electroosmotic flow control by an applied radial voltage for capillary zone electrophoresis. Anal. Chem. 1993, 65, 27–31.

35.

Gao

Lin

F. Y. H.

Sherman

P. M.

An electrokinetically-controlled immunoassay for simultaneous detection of multiple microbial antigens. Biomed. Microdevices 2005, 7(4), 301–312.

36.

Mugele

Baret

J. C.

Electrowetting: From basics to applications. J. Phys. Condens. Matter 2005, 17(28), R705–R774.

37.

Sista

R. S.

Eckhardt

A. E.

Srinivasan

Pollack

M. G.

Palanki

Pamula

V. K.

Heterogeneous immunoassays using magnetic beads on a digital microfluidic platform. Lab Chip 2008, 8(12), 2188–2196.

38.

Sista

R. S.

Hua

Thwar

Sudarsan

Srinivasan

Eckhardt

Pollack

Pamula

Development of a digital microfluidic platform for point of care testing. Lab Chip 2008, 8(12), 2091–2104.

39.

Wang

C. H.

Lee

G. B.

Automatic bio-sampling chips integrated with micro-pumps and micro-valves for disease detection. Biosens. Bioelectron. 2005, 21(3), 419–425.

40.

Wang

C. H.

Lee

G. B.

Pneumatically driven peristaltic micropumps utilizing serpentine-shape channels. J. Micromech. Microeng. 2006, 16(2), 341–348.

41.

Yang

S. Y.

Lien

K. Y.

Huang

K. J.

Lei

H. Y.

Lee

G. B.

Micro flow cytometry utilizing a magnetic bead-based immunoassay for rapid virus detection. Biosens. Bioelectron. 2008, 24(4), 855–862.

42.

Yang

Y. N.

Hsiung

S. K.

Lee

G. B.

A pneumatic micropump incorporated with a normally closed valve capable of generating a high pumping rate and a high back pressure. Microfluid. Nanofluidics 2009, 6(6), 823–833.

43.

Madou

Zoval

Jia

Kido

Kim

Lab on a CD. Annu. Rev. Biomed. Eng. 2006, 8, 601–628.

44.

Madou

Lee

L. J.

Daunert

Lai

Shih

C. H.

Design and fabrication of CD-like microfluidic platforms for diagnostics: Microfluidic functions. Biomed. Microdevices 2001, 3(3), 245–254.

45.

Norman

G. A.

Computer interfaced fast analyzers. Science 1969, 166(3903), 317–324.

46.

Nobuo

Ulrika

Per

Stephan

Hiroyuki

Simultaneous multiple immunoassays in a compact disc-shaped microfluidic device based on centrifugal force. Clin. Chem. 2005, 51(10), 1955–1961.

47.

Yuan

Wang

Chiou

N. R.

Epstein

A. J.

Lee

L. J.

Design and testing of a microfluidic biochip for cytokine enzyme-linked immunosorbent assay. Biomicrofluidics 2009, 3(2), 022401.

48.

Lee

B. S.

Lee

J. N.

Park

J. M.

Lee

J. G.

Kim

Cho

Y. K.

A fully automated immunoassay from whole blood on a disc. Lab Chip 2009, 9(11), 1548–1555.

49.

Hosokawa

Omata

Sato

Maeda

Power-free sequential injection for microchip immunoassay toward point-of-care testing. Lab Chip 2006, 6(2), 236–241.

50.

Kim

H. H.

Park

D. H.

Ryu

B. H.

Lim

K. J.

Design and modeling of piezoelectric pump for microfluid devices. Ferroelectr. 2009, 378(1), 92–100.

51.

S. M.

Cho

Ahn

Disposable thermo-pneumatic micropump for bio lab-on-a-chip application. Microelectron. Eng. 2009, 86(4–6), 1337–1339.

52.

Herrmann

Veres

Tabrizian

Enzymatically-generated fluorescent detection in micro-channels with internal magnetic mixing for the development of parallel microfluidic ELISA. Lab Chip 2006, 6(4), 555–560.

53.

Holmes

She

J. K.

Roach

P. L.

Morgan

Bead-based immunoassays using a micro-chip flow cytometer. Lab Chip 2007, 7(8), 1048–1056.

54.

Herrmann

Veres

Tabrizian

Quantification of lowpicomolar concentrations of TNF-α in serum using the dual-network microfluidic ELISA platform. Anal. Chem. 2008, 80(13), 5160–5167.

55.

Kevin

Heyriesa

K. A.

Mandonb

C. A.

Ceriottic

Pontic

Colpoc

Bluma

L. J.

Marquettea

C. A.

Macromolecules to PDMS transfer as a general route for PDMS biochips. Biosens. Bioelectron. 2009, 24(5), 1146–1152.

56.

Hashimoto

Kaji

Kemppinen

M. E.

Nishizawa

Localized immobilization of proteins onto microstructures within a preassembled microfluidic device. Sens. Actuators B Chem. 2008, 128(2), 545–551.

57.

Phillips

K. C.

Cheng

Microfluidic immunoassay for bacterial toxins with supported phospholipid bilayer membranes on poly(dimethylsiloxane). Anal. Chem. 2005, 77(1), 327–334.

58.

Dong

Phillips

K. S.

Cheng

Immunosensing of staphylococcus enterotoxin (SEB) in milk with PDMS microfluidic systems using reinforced supported bilayer membranes (rSBMs). Lab Chip 2006, 6(5), 675–681.

59.

C. M.

Liu

Y. S.

Gao

Wang

Gan

Flow-through functionalized PDMS microfluidic channels with dextran derivative for ELISAs. Lab Chip 2009, 9(9), 1243–1247.

60.

Jönsson

Aronsson

Rundström

Pettersson

Hartvig

Bakker

Martinsson

Liedberg

MacCraith

Ohmana

Melin

Silane-dextran chemistry on lateral flow polymer chips for immunoassays. Lab Chip 2008, 8(7), 1191–1197.

61.

Darain

Yager

Gana

K. L.

Tjin

S. C.

On-chip detection of myoglobin based on fluorescence. Biosens. Bioelectron. 2009, 24(6), 1744–1750.

62.

Darain

Gan

K. L.

Tjin

C. S.

Antibody immobilization on to polystyrene (PS) substrate-on-chip immunoassay for horse IgG based on fluorescence. Biomed. Microdevices 2009, 11(3), 653–661.

63.

Dandy

D. S.

Henry

C. S.

Microfluidic protein patterning on silicon nitride using solvent-extracted poly(dimethylsiloxane) channels. Sens. Actuators B Chem. 2008, 129(2), 811–817.

64.

Wang

Meng

Guo

Liu

Yang

P. Y.

Zhong

Liu

Microfluidic immunosensor based on stable antibody-patterned surface in PMMA microchip. Electrochem. Commun. 2008, 10(3), 447–450.

65.

Y. J.

Maeng

J. H.

Ahn

Hwang

S. Y.

Cho

N. G.

Lee

S. H.

Microchip-based multiplex electro-immunosensing system for the detection of cancer biomarkers. Electrophoresis 2008, 29(16), 3466–3476.

66.

Lee

N. Y.

Yang

Kim

Y. S.

Park

Microfluidic immunoassay platform using antibody-immobilized glass beads and its application for detection of Escherichia coli O157: H7. Bull. Korean Chem. Soc. 2006, 27(4), 479–483.

67.

Yang

Nam

S. W.

Lee

N. Y.

Kim

Y. S.

Park

Superporous agarose beads as a solid support for microfluidic immunoassay. Ultramicroscopy 2008, 108(10), 1384–1389.

68.

Sato

Tokeshi

Kimura

Kitamori

Determination of carcinoembryonic antigen in human sera by integrated bead-bed immunoassay in a microchip for cancer diagnosis. Anal. Chem. 2001, 73(6), 1213–1218.

69.

Sivagnanam

Song

Vandevyver

Gijs

M. A.

On-chip immunoassay using electrostatic assembly of streptavidin-coated bead micropatterns. Anal. Chem. 2009, 81(15), 6509–6515.

70.

Saunders

G. C.

Jett

J. H.

Martin

J. C.

Amplified flow-cytometric separation-free fluorescence immunoassays. Clin. Chem. 1985, 31(12), 2020–2023.

71.

Lindmo

Bormer

Ugelstad

Nustad

Immunometric assay by flow cytometry using mixtures of two particle types of different affinity. J. Immunol. Methods 1990, 126(2), 183–189.

72.

Frengen

Kierulf

Schmid

Lindmo

Nustad

Demonstration and minimization of serum interference in flow cytometric two-site immunoassays. Clin. Chem. 1994, 40(3), 420–425.

73.

Kuhara

Takeyama

Tanaka

Matsunaga

Magnetic cell separation using antibody binding with protein A expressed on bacterial magnetic particles. Anal. Chem. 2004, 76(21), 6207–6213.

74.

Yoza

Arakaki

Maruyama

Takeyama

Matsunaga

Fully automated DNA extraction from blood using magnetic particles modified with a hyperbranched polyamidoamine dendrimer. J. Biosci. Bioeng. 2003, 95(1), 21–26.

75.

Fan

Lau

Magnetic bead-based chemiluminescent metal immunoassay with a colloidal gold label. Anal. Chem. 2005, 77(10), 3238–3242.

76.

Lien

K. Y.

Lee

W. C.

Lei

H. Y.

Lee

G. B.

Integrated reverse transcription polymerase chain reaction systems for virus detection. Biosens. Bioelectron. 2007, 22(8), 1739–1748.

77.

Qiua

Zhoub

Chenb

Lina

J. M.

Immunomagnetic separation and rapid detection of bacteria using bioluminescence and microfluidics. Talanta 2009, 79(3), 787–795.

78.

Huang

Zheng

X. L.

Zheng

J. S.

Pan

X. Y.

Rapid analysis of alpha-fetoprotein by chemiluminescence microfluidic immunoassay system based on super-paramagnetic microbeads. Biomed. Microdevices 2009, 11(1), 213–216.

79.

Bunyakul

Edwards

K. A.

Promptmas

Baeumner

A. J.

Cholera toxin subunit B detection in microfluidic devices. Anal. Bioanal. Chem. 2009, 393(1), 177–186.

80.

Kim

K. S.

Park

J. K.

Magnetic force-based multiplexed immunoassay using superparamagnetic nanoparticles in microfluidic channel. Lab Chip 2005, 5(6), 657–664.

81.

Meagher

R. J.

Hatch

A. V.

Renzi

R. F.

Singh

A. K.

An integrated microfluidic platform for sensitive and rapid detection of biological toxins. Lab Chip 2008, 8(12), 2046–2053.

82.

Herr

A. E.

Hatch

A. V.

Throckmorton

D. J.

Tran

H. M.

Brennan

J. S.

Giannobile

W. V.

Singh

A. K.

Microfluidic immunoassays as rapid saliva-based clinical diagnostics. Proc. Natl. Acad. Sci. U.S.A. 2007, 104(13), 5268–5273.

83.

Reichmuth

D. S.

Wang

S. K.

Barrett

L. M.

Throckmorton

D. J.

Einfeld

Singha

A. K.

Rapid microchip-based electrophoretic immunoassays for the detection of swine influenza virus. Lab Chip 2008, 8(8), 1319–1324.

84.

Lin

C. C.

Hsu

B. K.

Chen

S. H.

Integrated isotachophoretic stacking and gel electrophoresis on a plastic substrate and variations in detection dynamic range. Electrophoresis 2008, 29(6), 1228–1236.

85.

Kawabata

Wada

H. G.

Watanabe

Satomura

Electrokinetic Analyte Transport Assay for α-fetoprotein immunoassay integrates mixing, reaction and separation on-chip. Electrophoresis 2008, 29(7), 1399–1406.

86.

Kagebayashi

Yamaguchi

Akinaga

Kitano

Yokoyama

Satomura

Kurosawa

Watanabe

Kawabata

Chang

Bousse

Wada

H. G.

Satomura

Automated immunoassay system for AFP-L3% using on-chip electrokinetic reaction and separation by affinity electrophoresis. Anal. Biochem. 2009, 388(2), 306–311.

87.

Lee

K. H.

Y. D.

Chen

S. J.

Tseng

F. G.

Lee

G. B.

Microfluidic systems integrated with two-dimensional surface plasmon resonance phase imaging systems for microarray immunoassay. Biosens. Bioelectron. 2007, 23(4), 466–472.

88.

Kurita

Yokota

Sato

Mizutani

Niwa

On-chip enzyme immunoassay of a cardiac marker using a microfluidic device combined with a portable surface plasmon resonance system. Anal. Chem. 2006, 78(15), 5525–5531.

89.

Kim

S. J.

Gobi

K. V.

Iwasaka

Tanaka

Miura

Novel miniature SPR immunosensor equipped with all-in-one multi-micro-channel sensor chip for detecting low-molecular-weight analytes. Biosens. Bioelectron. 2007, 23(5), 701–707.

90.

Tokeshi

Uchida

Hibara

Sawada

Kitamori

Determination of subyoctomole amounts of nonfluorescent molecules using a thermal lens microscope: Subsingle-molecule determination. Anal. Chem. 2001, 73(9), 2112–2116.

91.

Henares

T. G.

Funano

S. I.

Terabe

Mizutani

Sekizawa

Hisamoto

Multiple enzyme linked immunosorbent assay system on a capillary-assembled microchip integrating valving and immunoreaction functions. Anal. Chim. Acta 2007, 589(2), 173–179.

92.

Yamauchi

Tokeshi

Yamaguchi

Fukuzawa

Hattori

Hibara

Kitamori

Miniaturized thermal lens and fluorescence detection system for microchemical chips. J. Chromatogr. A 2006, 1106(1–2), 89–93.

93.

Cottier

Wiki

Voirin

Gao

Kunz

R. E.

Label-free highly sensitive detection of (small) molecules by wavelength interrogation of integrated optical chips. Sens. Actuators B Chem. 2003, 91(1–3), 241–251.

94.

Suárez

Jin

Y. H.

Auerswald

Berchtold

Knapp

H. F.

MarcDiserens

Leterrier

Månson

J. A. E.

Voirin

Lab-on-a-chip for multiplexed biosensing of residual antibiotics in milk. Lab Chip 2009, 9(11), 1625–1630.

95.

Liu

Zhang

Liua

Determination of β-lactam antibiotics in milk using micro-flow chemiluminescence system with on-line solid phase extraction. Anal. Chim. Acta 2007, 592(2), 187–192.

96.

Turnipseed

S. B.

Andersen

W. C.

Karbiwnyk

C. M.

Madson

M. R.

Miller

K. E.

Multi-class, multi-residue liquid chromatography/tandem mass spectrometry screening and confirmation methods for drug residues in milk. Rapid Commun. Mass Spectrom. 2008, 22(10), 1467–1480.

97.

Adrian

Pinacho

D. G.

Granier

Diserens

J. M.

Baeza

F. S.

Marco

M. P.

A multianalyte ELISA for immunochemical screening of sulfonamide, fluoroquinolone and β-lactam antibiotics in milk samples using class-selective bioreceptors. Anal. Bioanal. Chem. 2008, 391(5), 1703–1712.

98.

Knecht

B. G.

Strasser

Dietrich

Martlbauer

Niessner

Weller

M. G.

Automated microarray system for the simultaneous detection of antibiotics in milk. Anal. Chem. 2004, 76(3), 646–654.

99.

Taton

T. A.

Mirkin

C. A.

Letsinger

R. L.

Scanometric DNA array detection with nanoparticle probes. Science 2000, 289(5485), 1757–1760.

100.

Lin

F. Y. H.

Sabri

Alirezaie

Sherman

P. M.

Development of a nanoparticle-labeled microfluidic immunoassay for detection of pathogenic microorganisms. Clin. Diagn. Lab. Immunol. 2005, 12(3), 418–425.

101.

Shi

Qin

Lin

Low cost, portable detection of gold nanoparticle-labeled microfluidic immunoassay with camera cell phone. Electrophoresis 2009, 30(4), 579–582.

102.

Chew

F. T.

S. F. Y.

Design and application of piezoelectric quartz crystal-based immunoassay. Anal. Sci. 2000, 16(2), 107–114.

103.

Michalzik

Wilke

Büttgenbach

Miniaturized QCM-based flow system for immunosensor application in liquid. Sens. Actuators B Chem. 2005, 111–112(11), 410–415.

104.

Mulvaney

S. P.

Cole

C. L.

Kniller

M. D.

Malito

Tamanaha

C. R.

Rife

J. C.

Stanton

M. W.

Whitman

L. J.

Rapid, femtomolar bioassays in complex matrices combining microfluidics and magnetoelectronics. Biosens. Bioelectron. 2007, 23(2), 191–200.

105.

Zhao

X. W.

Sun

Zhu

Z. Z.

Encoded silica colloidal crystal beads as supports for potential multiplex immunoassay. Anal. Chem. 2008, 80(5), 1598–1605.

106.

Sun

Zhao

X. W.

Zhao

Y. J.

Zhu

Z. Z.

Fabrication of colloidal crystal beads by a drop-breaking technique and their application as bioassays. Small 2008, 4(5), 592–596.

107.

Matsunaga

Maeda

Yoshino

Takeyama

Takahashi

Ginya

Aasahina

Tajima

Fully automated immunoassay for detection of prostate-specific antigen using nano-magnetic beads and micro-polystyrene bead composites, ‘Beads on Beads’. Anal. Chim. Acta 2007, 597(2), 331–339.

108.

Hofmann

Wang

deMello

J. C.

Bradley

D. D. C.

deMello

A. J.

Towards microalbuminuria determination on a disposable diagnostic microchip with integrated fluorescence detection based on thin-film organic light emitting diodes. Lab Chip 2005, 5(8), 863–868.